JP2001157973A - Walk control device, and walk control method for robot - Google Patents

Abstract

PROBLEM TO BE SOLVED: To compensate or recover attitude stability lost for emotion expressing actions using the body or hands. SOLUTION: A total body motion pattern to realize stable walk is obtained by introducing a lower back motion pattern based on a desired foot part motion pattern, a ZMP orbit, a body trunk motion pattern, and an arm motion pattern. A walking mode of a leg is determined to make stable walk in any action condition of a robot including standing straight without loving, or walking. When action of the body or hands using the upper part of its body is applied when it is standing straight without moving, the walking mode of the legs is determined to make stable walk in accordance with the walking mode of the upper part of the body.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a walking control device and a walking control method for a realistic robot having a structure imitating the mechanism and operation of a living body, and more particularly to an upright walking type such as a human or monkey. The present invention relates to a walking control device and a walking control method for a legged mobile robot having a structure imitating a human body mechanism and operation.

More specifically, the present invention provides a leg-type movement by bipedal upright walking, and a torso and a head,
The present invention relates to a walking control device and a walking control method for an upright walking / legged mobile robot equipped with a so-called upper body such as an arm, and particularly relates to two feet while maintaining natural motion and rich expressiveness close to humans. The present invention relates to a walking control device and a walking control method for a robot that realizes stable walking by using a robot.

[0003]

2. Description of the Related Art A mechanical device that performs a motion similar to a human motion by using an electric or magnetic action is called a "robot". The robot is derived from the Slavic word "ROBO"
TA (slave machine) is said to be derived from robots. In Japan, robots began to spread from the late 1960s, but many of them were aimed at automation and unmanned production work in factories. They were industrial robots such as manipulators and transfer robots.

Recently, research and development on legged mobile robots that imitate the body mechanisms and movements of animals such as humans and monkeys that walk upright on two legs have progressed, and expectations for their practical use have increased. Leg-based movement by standing two feet is excellent in that a flexible running operation such as climbing up and down stairs and over obstacles can be realized.

[0005] The legged mobile robot has a history of starting from research on legged movement using only the lower limbs as an elemental technology, instead of a form equipped with all five standing bodies.

For example, Japanese Patent Laying-Open No. 3-184772 discloses a joint structure applied to a structure corresponding to a structure below a trunk of a legged walking robot.

Japanese Patent Application Laid-Open No. 5-305579 discloses a walking control device for a legged mobile robot. The walking control device described in the publication is a ZMP (Zero M
omentPoint), that is, a point on the floor where the moment due to the floor reaction force when walking becomes zero is controlled to coincide with the target value. However, as can be seen from FIG. 1 described in the publication, the body 24 acting a moment is formed as a black box, and not all five bodies are in a completed state. Stay.

However, the ultimate purpose of the legged mobile robot is, of course, the completion of a complete five-body structure. In other words, the lower limbs that walk bipedally, the torso and the head,
An object of the present invention is to perform an upright walking with two feet using a structure including an upper limb including an arm and a trunk connecting the lower limb and the upper limb. The robot with the five bodies completed is supposed to be upright and legged using two feet, and cooperates with the upper limb, lower limb, and torso according to the given priority in each scene of the work in the human living space. Need to be controlled.

In particular, a legged mobile robot that emulates a human mechanism or movement is called a "humanoid" or "humanoid" robot. The humanoid robot can perform, for example, life support, that is, support of human activities in various situations in a living environment and other daily lives.

[0010] Legged mobile robots can be broadly classified into those for industrial purposes as in the past and those for entertainment purposes.

[0011] The former robots for industrial purposes are used in industrial activities.
It is intended to realize various difficult tasks in production activities on behalf of humans. One example is the maintenance of nuclear power plants, thermal power plants, and petrochemical plants, and the substitution of dangerous and difficult operations such as in manufacturing plants and high-rise buildings. This type of robot is designed and manufactured to realize a specific application or function in the industry, and is premised on bipedal walking. It is not always necessary to faithfully reproduce a body mechanism or an operation mechanism as a mechanical device. For example, while enhancing the degree of freedom and movement functions of specific parts such as hands to realize specific applications, restricting or omitting the degrees of freedom of the head and waist, etc., which are relatively low in applications. Design convenience such as doing is provided. As a result, unnaturalness may remain in the appearance of the work or operation of the robot, even though it is called bipedal walking,
This has to be compromised.

On the other hand, the latter type of legged mobile robot for the purpose of entertainment has a property close to the life itself, rather than a life support for performing difficult tasks. That is, the ultimate purpose of this type of robot is to faithfully reproduce the mechanism inherent in an animal such as a human or a monkey walking upright with two legs, and to realize its natural smooth operation. Further, it is desirable that the entertainment robot has rich expressive power as long as it emulates an upright animal with high intelligence such as a human or a monkey. In this sense, entertainment robots that imitate humans are exactly what we call “humanoid robots”.

[0013] In short, even if it is a legged mobile robot, the robot for entertainment purposes shares the elemental technology with the industrial purpose robot, but on the other hand, the ultimate goal, the hardware mechanism and the operation for realizing it. It is no exaggeration to say that the control method and software configuration are completely different.

As is well known, the human body has hundreds of joints or hundreds of degrees of freedom. In order to give the legged mobile robot an action as close as possible to a human, it is preferable to give almost the same degree of freedom, but this is technically very difficult. It is necessary to arrange at least one actuator for each degree of freedom. However, mounting hundreds of actuators on a mechanical device such as a robot requires weight and weight in terms of manufacturing cost. It is almost impossible from a design point of view such as size. If the degree of freedom is large, the amount of calculation for position / motion control, balance control, and the like increases exponentially.

In summary, a humanoid robot must emulate a human body mechanism with limited degrees of freedom. For robots for entertainment purposes, there is a further requirement that natural movements closer to humans and expressive movements must be realized using far less degrees of freedom than the human body. is there.

A legged mobile robot performing bipedal upright walking is excellent in that it can realize a flexible running operation (for example, moving up and down stairs and over obstacles), but has a high center of gravity. The attitude control and the stable walking control become difficult to that extent. In particular, in the case of the entertainment type robot, the posture and stable walking must be controlled while richly expressing natural movements and emotions of intelligent animals such as humans and monkeys.

[0017]

By the way, when considering the “expressive power” of movements in humans and monkeys, the movements and postures of the upper body such as arms and torso not only realize the work but also express emotions. It has the aspect of embodiment and has a very important meaning.
This is why they are called “gesture” and “hand gesture”.

In daily life, gestures and hand gestures are performed during any period of the stationary state of the lower body, the walking and other movable states of the lower body. In addition, the position of the center of gravity of the entire body is largely moved by a gesture or a hand gesture, and a moment of inertia is generated. Real animals, such as humans and monkeys, are designed to autonomously compensate for the balance and moment of the center of gravity, and naturally perform the operation of maintaining a normal state of stopping or walking.

On the other hand, it is an important issue for humanoid robots to be rich in expressive power (described above), and gestures and hand gestures are indispensable. Therefore, adaptive posture control and stable walking control are required for upper body-led motions such as gestures and hand gestures.

A number of techniques relating to posture control and stable walking for a robot of the type that performs legged movement by biped walking have already been proposed in the art. However,
Many of these conventional proposals are based on ZMP (Zero Moment Poin
t), that is, a method of adaptively controlling a point on the floor where the moment due to the floor reaction force during walking becomes zero to coincide with a target value.

For example, the legged mobile robot described in Japanese Patent Application Laid-Open No. 5-305579 is designed to perform stable walking by matching a point on the floor at which ZMP becomes zero with a target value.

Further, in the legged mobile robot described in Japanese Patent Application Laid-Open No. Hei 5-305581, the ZMP is provided at least from the inside of the supporting polyhedron (polygon) or from the end of the supporting polyhedron (polygon) when landing or leaving the floor. It was configured to be at a position having a predetermined margin. As a result, there is room for the ZMP for a predetermined distance even when a disturbance or the like is received, and walking stability can be improved.

Japanese Patent Application Laid-Open No. 5-3055583 discloses that the walking speed of a legged mobile robot is controlled by a ZMP target position. That is, the legged mobile robot described in the publication uses the previously set walking pattern data, drives the leg joints so that the ZMP matches the target position, detects the inclination of the upper body, and The discharge speed of the walking pattern data set according to the detected value is changed. As a result, when the robot leans forward on unexpected irregularities, for example, the posture can be recovered by increasing the discharge speed.
Further, since the ZMP can be controlled to the target position, there is no problem even if the discharge speed is changed during the two-leg supporting period.

Japanese Patent Laid-Open Publication No. Hei 5-305585 discloses that a landing position of a legged mobile robot is controlled by a ZMP target position. That is, the legged mobile robot described in the publication detects a deviation between the ZMP target position and the actually measured position, and drives one or both of the legs so as to eliminate the deviation. Then, a stable walking is performed by detecting the moment and driving the legs so that the moment becomes zero.

Japanese Patent Laid-Open Publication No. Hei 5-305586 discloses that the inclination posture of a legged mobile robot is controlled by a ZMP target position. That is, the legged mobile robot described in the publication detects a moment around the ZMP target position, and when a moment is generated, drives the legs so that the moment becomes zero to perform stable walking. It has become.

Also, for example, “Biped Walking Robot Data Collection” (Revised 2nd Edition) Ministry of Education, Science and Technology Research Funds / Comprehensive Research (A)
"Research on Walking and Control of Biped Robots" Research Group (February 1986) and "Development of Biped Robots Compensating for 3-Axis Moment by Upper Body Motion" (6th Intelligent Mobile Robot Symposium, May 21, 1992, 2
On the 2nd) etc., a leg-type movement of a type in which at least an upper body joint that can be driven by the upper body is provided, a plurality of joints are provided on a leg connected to the upper body, and the leg joint is driven to walk. It is disclosed that in a robot, a gait of the upper body is determined based on a gait of a leg (that is, posture instability due to legged motion is compensated for by a gait of the upper body).

That is, the above-mentioned techniques do not generally perform posture control or stable walking control in consideration of an upper body-led operation. The above technique changes the gait (temporal movement) of the upper body when walking becomes difficult due to disturbance or the like during walking of a bipedal legged mobile robot, thereby stabilizing walking. It restores sex. In other words, the gait of the upper body is changed in order to compensate for instability due to disturbance during the operation led by the lower body, in other words, the expression of the upper body is sacrificed. Further, none of the legged mobile robots described in the above publications recovers a stable posture that has been lost due to an upper body-led emotion expression operation such as a gesture or a hand gesture.

The present invention has been made in view of the technical problems described above, and has as its object to provide an excellent walking control device and an excellent walking control device for a realistic robot having a structure imitating the mechanism and operation of a living body. It is to provide a walking control method.

It is a further object of the present invention to provide an excellent walking control device and method for a legged mobile robot having a structure simulating an upright walking type body mechanism or operation such as a human or monkey. To provide.

A further object of the present invention is to provide an upright walking / legged mobile type in which a so-called upper body such as a torso, a head, and an arm is mounted on a leg portion while performing a legged movement by bipedal upright walking. An object of the present invention is to provide an excellent walking control device and a walking control method for a robot.

A further object of the present invention is to provide an excellent gait control device and gait control method for a robot that realizes stable gait while maintaining natural motion and rich expressive power close to humans. is there.

A further object of the present invention is to provide a robot in which a so-called upper body such as a torso, a head, and an arm is mounted on a leg while performing a legged movement by bipedal upright walking. An object of the present invention is to provide an excellent gait control device and gait control method that can compensate or recover posture stability lost due to an upper body-led emotion expression operation.

It is a further object of the present invention to provide an excellent walking control device and a walking control for a two-leg upright walking type robot capable of determining a lower body gait capable of stably walking according to an upper body gait. The purpose of the present invention is to provide a control method ("gait" is a technical term meaning "chronological change of joint angle" in the art, and is substantially synonymous with "exercise pattern").

[0034]

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above problems, and a first aspect of the present invention comprises at least a lower limb, a trunk and a waist, and performs leg-type movement by the lower limb. A walking control device or a walking control method for a robot of any type, comprising:
A walking control apparatus or a walking control method for a robot, wherein a whole body movement pattern during walking is obtained by deriving a waist movement pattern based on a trunk movement pattern and an upper limb movement pattern.

According to a second aspect of the present invention, there is provided a walking control device or a walking control device for a robot of at least a type including a lower limb, a trunk and a waist, and performing legged movement with the lower limb so that the ZMP enters a target position. A method or a step for setting (a) a foot movement, a trunk movement, an upper limb movement, a posture and a height of a lumbar region for realizing a requested movement, and (b) the means or a step (b). means or steps for setting a ZMP trajectory based on the foot motion set according to a), and (c) means for obtaining a solution of a waist motion where moments are balanced on the ZMP set according to said means or step (b) or A walking control device or a walking control method for a robot, comprising: a step; and (d) means or a step of executing a waist movement based on a solution of the waist movement.

According to a third aspect of the present invention, there is provided a walking control device or a walking control device for a robot having at least a lower limb, a trunk and a waist, and performing legged movement with the lower limb so that the ZMP enters a target position. A method or a step for setting (A) a foot exercise, a trunk exercise, an upper limb exercise, a waist posture and a height for realizing a requested operation; and (B) the means or a step ( (A) means or a step for setting a ZMP trajectory based on the foot motion set by (A); and (C) a ZM set by the means or step (B) using an inexact model of the robot.
A means or a step for obtaining an approximate solution of the waist motion whose moment is balanced on P; and (D) a lumbar motion of which the moment is balanced on the ZMP set by the means or step (B) by using the strict model of the robot. A means or a step for obtaining an approximate solution; and (E) a means for calculating a waist motion if the difference between the approximate solutions of the means or the step (C) and the means or the step (D) is less than a predetermined allowable value. Or (F) the means or step (C);
And if the difference between the approximate solutions in the means or step (D) is equal to or greater than a predetermined allowable value, the moment of the inexact model on the set ZMP is corrected and re-input to the means (C). And (G) means or a step of executing a waist movement based on a solution of the waist movement.

In the robot walking control apparatus or method according to the third aspect of the present invention, the inexact model is, for example, a linear and / or non-interfering multi-mass point approximation model of the robot. Also, the strict model is, for example, a rigid body model related to the robot, or
A non-linear and / or interference multi-mass approximation model may be used. By using such an approximate model, the amount of calculation for walking control can be significantly reduced.

The robot walking control apparatus or method according to the third aspect of the present invention further comprises: (C ′)
If the means for obtaining an approximate solution of the waist movement using the inexact model or the approximate solution obtained in step (C) cannot realize the trunk / upper limb movement set in advance, resetting the trunk / upper limb movement pattern It may have means or steps for making corrections.

Further, in the robot walking control apparatus or the walking control method according to the third aspect of the present invention, the means or step (C) for obtaining an approximate solution of the waist motion using the inexact model is characterized in that: The approximate solution of the waist movement may be obtained by solving a balance equation between the moment on the set ZMP caused by the movement, trunk movement, and upper limb movement and the moment on the set ZMP caused by the movement in the horizontal plane of the waist. .

Alternatively, the means or step (C) for obtaining an approximate solution of the waist movement using the inexact model may be performed by replacing the time function with the frequency function.

Alternatively, the means or step (C) for obtaining an approximate solution of the waist movement using the above-described inexact model is a setting ZMP generated by a foot movement, a trunk movement, and an upper limb movement.
Apply the Fourier series expansion to the upper moment, apply the Fourier series expansion to the waist motion in the horizontal plane, calculate the Fourier coefficient of the trajectory in the waist horizontal plane, and apply the inverse Fourier series expansion to the waist motion. May be obtained.

A fourth aspect of the present invention is directed to an upper body having a plurality of joints for expressing the motion of the upper body and a lower body having joints for legs for realizing at least a walking motion. A walking control device or a walking control method for a walking type robot, comprising: determining a gait of a lower body such that a stable walking can be performed according to a gait of an upper body. It is a control method.

[0043]

According to the robot walking control apparatus and the walking control method of the present invention, it is possible to realize a waist motion capable of stable walking based on the settings of the trunk motion and the upper limb motion in addition to the foot motion. . The trunk movement and the upper limb movement correspond to an expression motion using the upper body of the robot, such as a gesture or a hand gesture, that is, a gait of the upper body.

That is, the gait of the lower limb can be determined so that stable walking can be realized even when the robot is in various operating states such as when the robot is standing upright or walking normally. In particular,
Gesture using upper body, upper limb and trunk, when standing upright
When a hand gesture is applied to the robot, the gait of the lower limb enables stable walking (ie, compensates for the balance lost due to the gait of the upper body) according to the gait of the upper body. Can be determined.

Still other objects, features and advantages of the present invention are:
It will become apparent from the following more detailed description based on the embodiments of the present invention and the accompanying drawings.

[0046]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIGS. 1 and 2 show the humanoid robot 100 used in the embodiment of the present invention viewed from the front and the rear, respectively. Further, FIG. 3 schematically shows the configuration of the degrees of freedom of the joints included in the humanoid robot 100.

As shown in FIG. 3, the humanoid robot 100
Is composed of an upper limb including two arms and a head 1, a lower limb including two legs for realizing a movement operation, and a trunk connecting the upper limb and the lower limb.

The neck joint that supports the head 1 includes a neck joint yaw axis 2, a neck joint pitch axis 3, and a neck joint roll axis 4.
It has a degree of freedom.

Each arm has a shoulder joint pitch axis 8, a shoulder joint roll axis 9, an upper arm yaw axis 10, an elbow joint pitch axis 11, a forearm yaw axis 12, a wrist joint pitch axis 13,
It is composed of a wrist joint roll shaft 14 and a hand 15. The hand portion 15 is actually a multi-joint / multi-degree-of-freedom structure including a plurality of fingers. However, the operation of the hand part 15 is
Since there is little contribution or influence on the posture control and the walking control of 0, in this specification, it is assumed that the degree of freedom is zero. Therefore,
Each arm has seven degrees of freedom.

The trunk has three degrees of freedom: a trunk pitch axis 5, a trunk roll axis 6, and a trunk yaw axis 7.

Each leg constituting the lower limb has a hip joint yaw axis 16, a hip joint pitch axis 17, a hip joint roll axis 18, a knee joint pitch axis 19, and an ankle joint pitch axis 2.
0, an ankle joint roll shaft 21, and a foot 22. In this specification, the intersection of the hip joint pitch axis 17 and the hip joint roll axis 18 defines the hip joint position of the robot 100 according to the present embodiment. Although the foot 22 of the human body is actually a structure including a sole with multiple joints and multiple degrees of freedom, the sole of the humanoid robot 100 according to the present embodiment has zero degrees of freedom.
Therefore, each leg has six degrees of freedom.

Summarizing the above, the humanoid robot 100 according to the present embodiment as a whole has a total of 3 + 7 × 2 + 3 +
It will have 6 × 2 = 32 degrees of freedom. However, the humanoid robot 100 for entertainment is not necessarily limited to 32 degrees of freedom. It goes without saying that the degree of freedom, that is, the number of joints, can be appropriately increased or decreased in accordance with design and manufacturing constraints, required specifications, and the like.

The degrees of freedom of the humanoid robot 100 as described above are actually implemented using actuators. It is preferable that the actuator is small and lightweight in view of requirements such as removing excess swelling on the appearance to approximate the human body shape and performing posture control on an unstable structure called bipedal walking. . In this embodiment,
A small AC servo actuator that is directly connected to the gear and built into the motor unit with the servo control system integrated into one chip is mounted. Note that this type of AC
The servo actuator is disclosed, for example, in Japanese Patent Application No. 11-33386 already assigned to the present applicant.

FIG. 4 schematically shows the configuration of a control system of the humanoid robot 100. As shown in the figure,
The humanoid robot 100 includes each of the mechanical units 30, 40, 50 R / L, and 60 R / L representing human limbs, and a control unit 80 that performs adaptive control for realizing a cooperative operation between the mechanical units. (However, each of R and L is a suffix indicating each of right and left. The same applies hereinafter).

The operation of the entire humanoid robot 100 is totally controlled by the control unit 80. The control unit 80 exchanges data and commands with a main control unit 81 composed of main circuit components (not shown) such as a CPU (Central Processing Unit) and a memory, and with a power supply circuit and each component of the robot 100. Peripheral circuit 82 including an interface (none of which is shown)
It is composed of

In realizing the present invention, the installation location of the control unit 80 is not particularly limited. In FIG. 4, it is mounted on the trunk unit 40, but may be mounted on the head unit 30. Alternatively, the control unit 80 may be provided outside the robot 100 to communicate with the robot 100 by wire or wirelessly.

Each joint degree of freedom in the robot 100 shown in FIG. 3 is realized by the corresponding actuator. That is, the head unit 30 includes a neck joint yaw axis actuator A ₂ , a neck joint pitch axis actuator A ₃ that expresses each of the neck joint yaw axis 2, the neck joint pitch axis 3, and the neck joint roll axis 4, and the neck joint roll. axis actuator A ₄ is disposed.

The trunk unit 40 includes a trunk pitch axis actuator A ₅ , a trunk roll axis actuator A ₆ that represents each of the trunk pitch axis 5, the trunk roll axis 6, and the trunk yaw axis 7, trunk yaw axis actuator A ₇ is deployed.

The arm unit 50R / L is subdivided into an upper arm unit 51R / L, an elbow joint unit 52R / L, and a forearm unit 53R / L. 9, upper arm yaw axis 10, elbow joint pitch axis 11, elbow joint roll axis 12, wrist joint pitch axis 1
3. Shoulder joint pitch axis actuator A ₈ , shoulder joint roll axis actuator A ₉ , upper arm yaw axis actuator A ₁₀ , elbow joint pitch axis actuator A ₁₁ , elbow joint roll axis actuator A ₁₂ representing each of the wrist joint roll axes 14 , A wrist joint pitch axis actuator A ₁₃ and a wrist joint roll axis actuator A ₁₄ are provided.

The leg unit 60R / L is subdivided into a thigh unit 61R / L, a knee unit 62R / L, and a shin unit 63R / L.
6. Hip joint yaw axis actuator A representing each of hip joint pitch axis 17, hip joint roll axis 18, knee joint pitch axis 19, ankle joint pitch axis 20, and ankle joint roll axis 21
₁₆ , a hip joint pitch axis actuator A ₁₇ , a hip joint roll axis actuator A ₁₈ , a knee joint pitch axis actuator A ₁₉ , an ankle joint pitch axis actuator A ₂₀ , and an ankle joint roll axis actuator A ₂₁ .

The actuators A ₂ ,
A ₃ ... Are more preferably small AC servo actuators (described above) of the type directly connected to a gear and of a type in which the servo control system is integrated into one chip and mounted in the motor unit.

The head unit 30, the trunk unit 40,
For each mechanism unit such as the arm unit 50 and each leg unit 60, a sub-control unit 3 for actuator drive control is provided.
5, 45, 55 and 65 are provided. Further, grounding confirmation sensors 91 and 92 for detecting whether or not the soles of the legs 60R and L have landed are mounted, and a posture sensor 93 for measuring a posture is provided in the trunk unit 40. ing.

The main control unit 80 performs adaptive control on the sub-control units 35, 45, 55 and 65 in response to the outputs of the sensors 91 to 93, and controls the upper limbs, trunk, And realize the coordinated movement of the lower limbs. Main controller 81
Is a foot exercise, ZMP according to a user command, etc.
(Zero Moment Point) The trajectory, the trunk movement, the upper limb movement, the waist height, and the like are set, and commands for instructing the operation according to the set contents are issued to the sub-control units 35, 45,
Transfer to 55 and 65. Then, each sub-control unit 35,
45, interpret the received command from the main control unit 81 and output a drive control signal to each of the actuators A ₂ , A ₃ . Here, “ZMP” refers to a point on the floor where the moment due to the floor reaction force during walking becomes zero, and “ZMP trajectory” refers to, for example, the robot 100
Means the trajectory of the movement of the ZMP during the walking operation period of FIG.

In this embodiment, the humanoid robot 100 physically having the multi-joint degree of freedom configuration shown in FIG. 3 is further replaced with a multi-mass point approximation model to perform the arithmetic processing of the walking control. . Real humanoid robot 100
Is a collection of infinite or continuous mass points, but its main purpose is to reduce the amount of computation by replacing it with an approximate model consisting of a finite number of discrete mass points.

FIG. 5 illustrates a linear and non-interfering multi-mass point approximation model of the humanoid robot 100 introduced for the calculation of the walking control according to the present embodiment.

In FIG. 5, the O-XYZ coordinate system represents the roll, pitch, and yaw axes in the absolute coordinate system.
The O'-X'Y'Z 'coordinate system represents the roll, pitch, and yaw axes in the motion coordinate system that moves with the robot 100. The multi-mass point model shown in FIG, i is a subscript representing a mass given to i-th, m _i is the i-th material point mass, r _'i is the i-th material point of the position vector (where motion coordinates System). In addition, the mass of the lumbar mass, which is particularly important in the lumbar motion control described later, is m _h , and its position vector is r ′ _h (r ′ _hx , r ′ _hy , r ′ _hz ),
Also, the position vector of the ZMP is r ′ _zmp .

In the inexact multi-mass point approximation model shown in FIG. 5, the moment equation is described in the form of a linear equation.
It should be appreciated that the moment equation does not interfere with the pitch and roll axes.

Such a multi-mass point approximation model can be generally generated by the following processing procedure.

(1) The mass distribution of the entire robot 100 is obtained. (2) Set a mass point. The method of setting the mass may be either manual input by a designer or automatic generation according to a predetermined rule. (3) for each of the regions i, obtains the center of gravity, is applied to the mass to the appropriate position of the center of gravity and mass m _i. (4) the mass points m _i, centered on the mass point position r _i, to display as a sphere having a radius proportional to its mass. (5) The masses that are actually connected, that is, the spheres are connected.

The multi-mass model is a representation of a robot in the form of a wire frame model. In the present embodiment, as can be seen from FIG. 5, this multi-mass point approximation model sets each of both shoulders, both elbows, both wrists, trunk, waist, and both ankles as mass points. .

The waist information of the multi-mass model shown in FIG.
At each rotation angle (θ_hx, Θ_hy, Θ _hz) Is a humanoid robot
Posture of the waist in the unit 100, that is, roll,
This defines the rotation of the hive and yaw axes.
The enlarged view around the waist of the mass model is shown.
I want to.)

Next, the humanoid robot 1 according to the present embodiment
The walking control processing procedure at 00 will be described.

Normally, a robot performs a predetermined operation by driving and controlling each joint, that is, an actuator according to a motion pattern generated in advance before the operation. In the case of the robot 100 according to the present embodiment, a waist movement pattern enabling stable walking is generated based on an arbitrary foot movement pattern, a ZMP trajectory, a trunk movement pattern, an upper limb movement pattern, and the like. . Here, the ZMP (Zero Moment Point) trajectory is a point at which a moment does not occur during a walking motion when the sole (or sole) of the walking robot is fixed to the floor at a certain point. (Described above).

As in this embodiment, one foot has six degrees of freedom.
In the case of a foot-walking robot (see FIG. 3), the posture of both legs is uniquely determined by the position of each foot 22R / L and the horizontal position and height of the waist. Therefore, generating the waist movement pattern is based on the posture of the legs, ie, the “gait” of the lower limbs
("Gait" is a technical term meaning "chronological change of joint angle" in the art, and is substantially synonymous with "exercise pattern").

FIG. 6 shows a robot 100 according to this embodiment.
5 shows a processing procedure for controlling a lumbar movement capable of walking stably in the form of a flowchart. However, in the following, each joint position and motion of the robot 100 will be described using a linear / non-interfering multi-mass point approximation model as shown in FIG. 5, and the following parameters will be used in the calculation. . However, it should be understood that symbols with a dash (') describe the motion coordinate system.

[0077]

(Equation 1)

[0078] Further, at a certain waist height of the robot 100 _{(r 'hz + r qz =} const), and assumes that knee mass point is zero.

The processing procedure is started in response to an input of a user command or the like for instructing the robot 100 to perform an action such as walking, gesturing or hand gesture. The operation of the robot 100 instructed by the user command includes, for example, a gesture / hand gesture using the upper limb and trunk when standing upright and immobile, a gesture using the upper limb and trunk during biped walking, There are various gestures.

The user command is sent to the main control unit 81
Interpreted in the foot (more specifically, sole) movement,
A pattern for actually determining the drive and operation of each part, such as the ZMP trajectory derived from the foot movement, the trunk movement, the upper limb movement, the posture and height of the waist, and the like is set (step S1).
1). More specifically, a foot motion pattern, a ZMP trajectory, a trunk motion pattern, and an upper limb motion pattern are set first. Also, regarding the motion of the waist, only the Z ′ direction is set, and the X ′ and Y ′ directions are unknown.

Next, a linear / non-interfering multi-mass point approximation model is used.
And settings caused by foot, trunk, and upper limb movement
Each moment about pitch axis and roll axis on ZMP
(M _x, M_y) Is calculated (step S12).

Next, using the linear / non-interfering multi-mass point approximation model, the moment on the set ZMP generated by the motion in the waist horizontal plane (r ′ _hx , r ′ _hy ) is calculated (step S13).

Next, a balance equation relating to the moment on the set ZMP is derived on the motion coordinate system O'-X'Y'Z 'that moves with the robot (step S14). More specifically, foot, trunk, and moments generated by the upper limb _{(M x,} M _y) on the right side as terms of known variables, section on horizontal movement of the lumbar mass point (r _hx, r _hy) of A linear and non-interfering ZMP equation (1) as shown in the following equation is derived as a term of the unknown variable on the left side.

[0084]

(Equation 2)

However, it is assumed that the following holds.

[0086]

(Equation 3)

Next, the above-mentioned ZMP equation (1) is solved to calculate the trajectory in the waist horizontal plane (step S15).
For example, by solving the ZMP equation (1) using a numerical solution (well-known) such as the Euler method or the Runge-Kutta method,
A numerical solution of the horizontal absolute position ( _rhx , _rhy ) of the waist as an unknown variable can be obtained (step S16). The numerical solution obtained here is an approximate solution of the waist movement pattern that enables stable walking, and more specifically, the waist horizontal absolute position where the ZMP enters the target position. The ZMP target position is usually set to the sole that has landed.

If the predetermined trunk / upper limb movement cannot be realized on the calculated approximate solution, the trunk / upper limb movement pattern is reset / corrected (step S17). At this time, the trajectory of the knee may be calculated.

Next, the strict model (that is, rigid body,
Alternatively, a moment (eM) on the set ZMP in the robot 100 (a precise model of a large number of mass points)
_x, eM _y) is calculated (step S18). The non-strict model assumes that the above [Equation 3] holds, but the strict model does not require such a premise (that is, it does not need to be constant with time).

The moment (eM) in the exact model_x,
eM_y) Is the moment error caused by waist movement.
You. In the following step S19, this moment (e
M_x, EM _y) Is the approximate moment in the inexact model
Tolerance (εM_x, ΕM_y) Is determined. Forgiveness
Exact solution of waist stable motion pattern if less than ε
To obtain a whole-body exercise pattern that can achieve stable walking
This means that the processing has been completed (step S20).
End the entire routine.

On the other hand, the moment (e
If M _x , eM _y ) is greater than or equal to the allowable value of the moment (εM _x , εM _y ) in the approximate model, the known moment (eM _x , eM _y ) in the approximate model is used using the moment (eM _x , eMy) in the strict model. M _x, modify the _{M y)} (step S21), and performs the derivation of ZMP equation once again. Then, the calculation and correction of the approximate solution of the waist motion pattern as described above are repeatedly executed until the value converges to less than the allowable value ε.

In short, according to the processing procedure shown in FIG.
In other words, it is possible to realize a lumbar motion capable of stable walking based on the settings of the trunk motion and the upper limb motion in addition to the leg motion. The trunk movement and the upper limb movement correspond to an expression operation using the upper body of the robot 100, such as a gesture or a hand gesture, that is, a gait of the upper body. In the case of a bipedal walking robot 100 having six degrees of freedom on one leg (see FIG. 3), the position of each leg 22R / L, the horizontal position and height of the waist, and the posture of the leg, that is, the leg Since the “gait” is uniquely determined, generating the waist movement pattern means determining the “gait” of the lower limb.

Therefore, according to the two-legged upright walking type robot 100 according to the present embodiment, the gait of the lower limb can be realized so that stable walking can be realized even in various operating states such as when standing upright immovably or during normal walking. Can be paraphrased.

In particular, when the upper body, that is, a gesture or hand gesture using the upper limb and the trunk is applied to the robot 100 when the robot is standing upright, the robot can walk stably in accordance with the gait of the upper body (ie, The gait of the lower limb (to compensate for the imbalance associated with the upper gait) can be determined. In other words, postural stability lost due to upper body-led emotional expression movements such as gestures and hand gestures can be suitably compensated or recovered by the gaits of the lower limbs defined by the horizontal position of the waist. .

FIG. 7 is a flowchart showing another example of the processing procedure for controlling the lumbar movement enabling stable walking in the robot 100 according to the present embodiment. However, in the same manner as described above, each joint position and motion of the robot 100 are described using a linear / non-interfering multi-mass point approximation model.

The processing procedure is started in response to an input of a user command or the like for instructing the robot 100 to perform an operation such as walking, gesturing or hand gesture. The operation of the robot 100 instructed by the user command includes, for example, a gesture / hand gesture using the upper limb and trunk when standing upright and immobile, a gesture using the upper limb and trunk during biped walking, There are various gestures.

The user command is sent to the main control unit 81
Interpreted in the foot (more specifically, sole) movement,
A pattern for actually determining the drive and operation of each part, such as the ZMP trajectory derived from the foot movement, the trunk movement, the upper limb movement, and the posture and height of the waist, is set (step S3).
1). More specifically, a foot motion pattern, a ZMP trajectory, a trunk motion pattern, and an upper limb motion pattern are set first. Also, regarding the motion of the waist, only the Z ′ direction is set, and the X ′ and Y ′ directions are unknown.

Next, using the linear / non-interfering multi-mass point approximation model (see above and FIG. 5), the pitch axis on the set ZMP generated by the foot, trunk, and upper limb movements,
Each moment about the roll axis (M _x, M _y) is calculated (step S32).

Next, the motion in the waist horizontal plane (r ′ _hx ,
r ′ _hy ) is _subjected to Fourier series expansion (step S33).
As is well known in the art, by performing Fourier series expansion, it is possible to replace the time axis component with the frequency component and perform the calculation. That is, in this case, the movement of the waist can be regarded as a periodic movement. Also, F
Since FT (fast Fourier transform) can be applied, the calculation speed can be greatly improved.

Next, the moments (M _x , M _y ) about the pitch axis and roll axis on the set ZMP are also subjected to Fourier series expansion (step S34).

Next, the approximate solution of the waist motion is obtained by calculating the Fourier coefficient of the trajectory in the horizontal plane of the lumbar region and further developing the inverse Fourier series (step S35).
36). The approximate solution obtained here is an approximate solution (r _hx , _rhy ) of the horizontal absolute position of the waist defining a waist movement pattern capable of stable walking, and more specifically, the waist such that the ZMP enters the target position. The absolute horizontal position. The ZMP target position is usually set to the sole that has landed.

If the previously set trunk / upper limb movement cannot be realized on the calculated approximate solution, the trunk / upper limb movement pattern is reset / corrected (step S37). At this time, the trajectory of the knee may be calculated.

Next, by substituting the whole body movement pattern obtained as described above, an exact model (that is, a rigid body,
Alternatively, a moment (eM) on the set ZMP in the robot 100 (a precise model of a large number of mass points)
_x, eM _y) is calculated (step S38). The non-strict model assumes that the above [Equation 3] holds, but the strict model does not require such a premise (that is, it does not need to be constant with time).

The moment (eM) in the exact model_x,
eM_y) Is the moment error caused by waist movement.
You. In a succeeding step S39, this moment (e
M_x, EM _y) Is the tolerance of the moment in the approximate model
Value (εM_x, ΕM_y) Is determined. Tolerance ε
If less than, the exact solution and stability of the lumbar stable motion pattern
I was able to obtain a whole body movement pattern that can realize walking
(Step S40), this processing routine
Terminate the entire program.

On the other hand, the moment (e
If M _x , eM _y ) is equal to or greater than the allowable value of the moment (εM _x , εM _y ) in the approximate model, the known generated moment in the inexact model is calculated using the moment (eM _x , eM _y ) in the exact model. (M _{x, M y)} by modifying the (step S41), and Fourier series expansion again, to converge to less than the allowable value epsilon, repeatedly executes the correction and calculation of the approximate solution of the waist motion pattern as described above .

Those skilled in the art will understand that the processing procedure shown in FIG. 7 can also realize a waist movement that enables stable walking based on the settings of the trunk movement and the upper limb movement. In particular, the Fourier series expansion can be used at high speed without resorting to the numerical solution of the ZMP equation, and at the same time, the moment calculation itself can be speeded up by applying FFT (Fast Fourier Transform). .

The trunk movement and the upper limb movement correspond to an expression operation using the upper body of the robot, such as a gesture or a hand gesture, that is, a gait of the upper body. In the case of a bipedal walking robot 100 having six joints with one leg (see FIG. 3), the position of each leg 22R / L and the height of the waist are uniquely determined by the leg posture. Generating the waist movement pattern means determining the posture of the leg, that is, the “gait” of the lower limb (described above).

Therefore, according to the robot 100 of the two-legged upright walking type according to the present embodiment, the gait of the lower limb can be realized so that stable walking can be realized even in various operating states such as standing upright immobility and normal walking. Can be paraphrased.

In particular, when a gesture or hand gesture using the upper body, that is, the upper limb and the trunk, is applied when the robot is standing upright and immobile,
The gait of the lower limb that enables stable walking can be determined according to the gait of the upper body. In other words, postural stability lost due to upper body-led emotional expression movements such as gestures and hand gestures can be suitably compensated or recovered by the gaits of the lower limbs defined by the horizontal position of the waist. .

[Supplement] The present invention has been described in detail with reference to the specific embodiments. However, it is obvious that those skilled in the art can modify or substitute the embodiment without departing from the spirit of the present invention.

In the present specification, the three degrees of freedom of the trunk pitch axis 5, the trunk roll axis 6, and the trunk yaw axis 7 of the robot are defined by the posture of the waist of the robot (θ _hx , θ _hy , θ _hz ). However, the position of the waist should be interpreted flexibly by comparing the humanoid robot 100 with the body mechanism of a real bipedal walking animal such as a human or a monkey.

The gist of the present invention is not necessarily limited to a product called a “robot”. That is, as long as the mechanical device performs a motion similar to a human motion using an electric or magnetic action, the present invention similarly applies to a product belonging to another industrial field such as a toy. Can be applied.

In short, the present invention has been disclosed by way of example, and should not be construed as limiting.
In order to determine the gist of the present invention, the claims described at the beginning should be considered.

For reference, a joint model configuration of a humanoid robot is illustrated in FIG. In the example shown in the figure, a portion consisting of the shoulder joint 5 to the upper arm 3, the elbow joint 6, the forearm 4, the wrist joint 7, and the hand 8 is referred to as an "upper limb" 17. In addition, the range from the shoulder joint 5 to the trunk joint 10 is referred to as “trunk” 9,
It corresponds to the human torso. Further, a range from the hip joint 11 to the trunk joint 10 is referred to as a “waist” 18. Trunk joint 10
Has the effect of expressing the degrees of freedom of the human spine.
In addition, a portion including the thigh 12, the knee 14, the lower leg 13, the ankle 15 and the foot 16 below the hip 11 is referred to as a “lower limb” 19. Generally, the upper part of the trunk joint 10 is called “upper body (or upper body)”, and the lower part is called “lower body (or lower body)”.

FIG. 10 illustrates another joint model configuration of the humanoid robot. The example shown in FIG. 11 differs from the example shown in FIG.
See the figure for the names of each part. As a result of omitting the trunk joint corresponding to the spine, the movement of the upper body of the humanoid robot loses its expressive power. However, in the case of a humanoid robot for industrial purposes, such as a dangerous operation or a surrogate operation, there is a case where the body does not need to move. It should be understood that the reference numerals used in FIGS. 9 and 10 do not match those of the other drawings.

[0116]

As described above in detail, according to the present invention,
It is possible to provide an excellent walking control device and an excellent walking control method that are suitably applied to a realistic robot having a structure imitating a biological mechanism or operation.

Further, according to the present invention, there is provided an excellent walking control device and an excellent walking control which are suitably applied to a legged mobile robot having a structure simulating an upright walking type body mechanism or operation such as a human or monkey. A control method can be provided.

Further, according to the present invention, an upright walking / legged mobile type in which a so-called upper body such as a torso, a head, and an arm is mounted on a leg portion while performing a legged movement by bipedal upright walking. An excellent walking control device and an excellent walking control method that are suitably applied to a robot can be provided.

Further, according to the present invention, there is provided an excellent gait control device and an excellent gait control method which are suitably applied to a robot which realizes a stable gait while maintaining a natural motion and rich expressive power close to humans. can do.

According to the present invention, there is provided a robot in which a so-called upper body such as a torso, a head, and an arm is mounted on a leg portion while performing a legged movement by bipedal upright walking. It is possible to provide an excellent walking control device and an excellent walking control method, which can compensate or recover the posture stability lost due to the upper body-led emotion expression operation.

The bipedal upright walking type robot according to the present invention generates a waist movement pattern enabling stable walking based on an arbitrary foot movement pattern, ZMP trajectory, trunk movement pattern, upper limb movement pattern and the like. As a result, a stable exercise pattern of the trunk or lower part (or waist or lower part), that is, the lower body, is generated. Therefore, the posture stability lost due to the upper body-led emotional expression operation such as gesture or hand gesture can be suitably compensated or restored by the exercise of the lower body.

In the case of a bipedal walking robot having six joint degrees of freedom on one leg, the posture of the leg is uniquely determined by the position of each leg and the height of the waist. This means determining the posture of the legs, that is, the “gait” of the lower limbs. Therefore, according to the gait control device and the gait control method according to the present invention, it is possible to determine a gait of a lower body that allows a stable gait according to a gait of an upper body.

[Brief description of the drawings]

FIG. 1 shows a humanoid robot 100 used for carrying out the present invention.
It is the figure which showed a mode that looked at from the front.

FIG. 2 is a diagram illustrating a humanoid robot 100 according to an embodiment of the present invention.
It is the figure which showed the mode that looked at from the back.

FIG. 3 is a diagram schematically illustrating a degree-of-freedom configuration model included in the humanoid robot 100 according to the present embodiment.

FIG. 4 is a diagram schematically illustrating a control system configuration of the humanoid robot 100 according to the present embodiment.

FIG. 5 is a diagram illustrating a linear and non-interfering multi-mass point approximation model of the humanoid robot 100, which is introduced for the calculation of walking control according to the present embodiment.

FIG. 6 is a flowchart showing a processing procedure of walking control in the humanoid robot 100 according to the embodiment.

FIG. 7 is a flowchart illustrating another example of a processing procedure for controlling a lumbar motion capable of stably walking in the robot 100 according to the embodiment.

FIG. 8 is an enlarged view of the periphery of the waist of the multi-mass model shown in FIG. 5;

FIG. 9 is a diagram schematically illustrating a joint model configuration of a humanoid robot.

FIG. 10 is a diagram schematically illustrating a joint model configuration of a humanoid robot.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Head, 2 ... Neck joint yaw axis 3 ... Neck joint pitch axis, 4 ... Neck joint roll axis 5 ... Trunk pitch axis, 6 ... Trunk roll axis 7 ... Trunk yaw axis, 8 ... Shoulder joint pitch axis 9 ... shoulder joint roll axis, 10 ... upper arm yaw axis 11 ... elbow joint pitch axis, 12 ... forearm yaw axis 13 ... wrist joint pitch axis, 14 ... wrist joint roll axis 15 ... hand, 16 ... hip joint yaw axis 17 ... hip joint Pitch axis, 18: Hip joint roll axis 19: Knee joint pitch axis, 20: Ankle joint pitch axis 21: Ankle joint roll axis, 22: Foot 30: Head unit, 40: Trunk unit 50: Arm unit, 51 ... upper arm unit 52 ... elbow joint unit, 53 ... forearm unit 60 ... leg unit, 61 ... thigh unit 62 ... knee joint unit, 63 ... shin unit 80 ... control unit, 81 ... main control unit 82 ... peripheral circuit 91, 2 ... ground check sensor 93 ... attitude sensor 100 ... humanoid robot

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Kenzo Ishida 6-7-35 Kita-Shinagawa, Shinagawa-ku, Tokyo Inside Sony Corporation (72) Inventor Jinichi Yamaguchi 5-14-38 Tamadaira, Hino-shi, Tokyo F term (reference) 2C150 CA01 CA02 DA02 DK02 DK03 EB01 ED01 ED08 ED41 EF16 EF23 EH07 FA01 3F059 AA00 BB06 DA07 FC00

Claims (18)

[Claims] 1. A walking control device for a robot comprising at least a lower limb, a trunk and a waist, and performing a legged movement by the lower limb, comprising: an arbitrary foot motion pattern, a ZMP trajectory, a trunk motion pattern, an upper limb A walking control device for a robot, wherein a whole body movement pattern during walking is obtained by deriving a waist movement pattern based on the movement pattern. 2. A walking control device for a robot of a type comprising at least a lower limb, a trunk and a waist, and performing leg-type movement with the lower limb so that the ZMP enters a target position,
(A) means for setting a foot movement, trunk movement, upper limb movement, waist posture and height for realizing a requested operation; and (b) foot movement set by the means (a). Means for setting a ZMP trajectory based on the following equation: (c) means for obtaining a solution of a waist motion in which moments are balanced on the ZMP set by the means (b); and (d) waist motion based on a solution of the waist motion. And a means for performing the following. 3. A walking control device for a robot of a type comprising at least a lower limb, a trunk and a waist, and performing legged movement with the lower limb so that the ZMP enters a target position,
(A) means for setting a foot movement, trunk movement, upper limb movement, waist posture and height for realizing a requested operation; and (B) foot movement set by the means (A). Means for setting a ZMP trajectory based on the following formula: (C) means for obtaining an approximate solution of waist motion in which moments are balanced on the ZMP set by means (B), using an inexact model of the robot; D) means for obtaining an approximate solution of waist movement in which moments are balanced on the ZMP set by means (B) by using the exact model of the robot; and (E) means for obtaining means (C) and means (D). If the difference between the approximate solutions is less than a predetermined allowable value, a means for determining a waist motion solution; and (F) the difference between the approximate solutions of the means (C) and the means (D) is equal to or greater than a predetermined allowable value. If, for example, the non-strict model Correct bets, said means (C)
And (G) means for performing a waist motion based on a solution of the waist motion. 4. The non-rigid model is a linear and / or non-interfering multi-mass point approximation model for the robot, and the strict model is a rigid body model or a non-linear and / or interference multi-mass point approximation model for the robot. is there,
The walking control device for a robot according to claim 3, wherein: 5. The method according to claim 5, further comprising: (C ') means for obtaining an approximate solution of waist movement using said inexact model; 4. The walking control device for a robot according to claim 3, further comprising: means for resetting / correcting a trunk / upper limb movement pattern. 6. The means (C) for obtaining an approximate solution of the waist motion using the inexact model comprises: a moment on a set ZMP caused by a foot motion, a trunk motion, and an upper limb motion; 4. The robot walking control device according to claim 3, wherein an approximate solution of the waist motion is obtained by solving a balance equation with a moment on a set ZMP generated by the robot. 7. The robot according to claim 3, wherein said means (C) for calculating an approximate solution of the waist motion using said inexact model replaces a time function with a frequency function. Walking control device. 8. The means (C) for obtaining an approximate solution of the waist motion using the inexact model applies Fourier series expansion to the moment on the set ZMP caused by the foot motion, the trunk motion, and the upper limb motion. Along with applying the Fourier series expansion to the waist motion in the horizontal plane, calculating the Fourier coefficient of the trajectory in the waist horizontal plane, and further calculating the approximate solution of the waist motion by applying the inverse Fourier series expansion, The walking control device for a robot according to claim 3. 9. A walking control for a walking robot comprising an upper body having a plurality of joints for expressing the motion of the upper body and a lower body having leg joints for realizing at least the walking motion. Device, depending on the gait of the upper body,
A walking control device for a robot, which determines a gait of a lower body such that a stable walking can be performed. 10. A walking control method for a robot of a type comprising at least a lower limb, a trunk and a waist, and performing legged movement with the lower limb, comprising: an arbitrary foot movement pattern, a ZMP trajectory, a trunk movement pattern, an upper limb A walking control method for a robot, wherein a whole body movement pattern during walking is obtained by deriving a waist movement pattern based on the movement pattern. 11. A walking control method for a robot of a type comprising at least a lower limb, a trunk and a waist, and performing a legged movement with the lower limb so that the ZMP enters a target position,
(A) setting a foot exercise, a trunk exercise, an upper limb exercise, a waist posture and a height for realizing a requested operation; and (b) a foot exercise set by the step (a). Setting a ZMP trajectory based on
(C) obtaining a solution of a waist motion in which moments are balanced on the ZMP set in the step (b); and (d) executing a waist motion based on the solution of the waist motion. Characteristic robot walking control method. 12. A walking control method for a robot of a type comprising at least a lower limb, a trunk and a waist, and performing leg-type movement with the lower limb so that the ZMP enters a target position,
(A) a step of setting a foot exercise, a trunk exercise, an upper limb exercise, a waist posture and a height for realizing a requested operation; and (B) a foot exercise set by the step (A). Setting a ZMP trajectory based on
(C) using an inexact model of the robot to obtain an approximate solution of the waist motion in which moments are balanced on the ZMP set in step (B); and (D) using an exact model of the robot, Step (B)
Calculating an approximate solution of the waist motion in which the moments are balanced on the ZMP set by the formula (E); and (E) the waist motion if the difference between the approximate solutions of the steps (C) and (D) is less than a predetermined allowable value. (F)
If the difference between the approximate solutions in the step (C) and the step (D) is equal to or more than a predetermined allowable value, the moment of the inexact model on the set ZMP is corrected, and the step is returned to the step (C). And (G) performing a waist motion based on a solution of the waist motion. 13. The non-strict model is a linear and / or non-interfering multi-mass point approximation model for the robot, and the strict model is a rigid body model or a non-linear and / or interference multi-mass point approximation model for the robot. is there,
The method for controlling walking of a robot according to claim 12, wherein: 14. The method according to claim 1, further comprising the step of: (C ') obtaining an approximate solution of waist movement using said inexact model; 13. The walking control method for a robot according to claim 12, further comprising the step of resetting and correcting a trunk / upper limb movement pattern. 15. The step (C) of obtaining an approximate solution of the waist motion using the inexact model includes the moment on the set ZMP caused by the foot motion, the trunk motion, and the upper limb motion, and the motion of the waist in the horizontal plane. The robot walking control method according to claim 12, wherein an approximate solution of the waist motion is obtained by solving a balance equation with a moment on a set ZMP generated by the robot. 16. The method according to claim 1, wherein the step (C) of obtaining an approximate solution of the waist motion using the inexact model is performed by replacing a function of time with a function of frequency.
3. The walking control method for a robot according to 2. 17. The step (C) of obtaining an approximate solution of waist motion using the inexact model applies Fourier series expansion to moments on a set ZMP caused by foot motion, trunk motion, and upper limb motion. Along with applying the Fourier series expansion to the motion of the waist in the horizontal plane, calculating the Fourier coefficient of the trajectory in the waist horizontal plane, and further calculating the approximate solution of the waist motion by applying the inverse Fourier series expansion,
The method for controlling walking of a robot according to claim 12, wherein: 18. A walking control for a walking robot comprising an upper body having a plurality of joints for expressing the movement of the upper body and a lower body having a joint of legs for realizing at least a walking operation. A method for controlling walking of a robot, comprising: determining a gait of a lower body capable of stably walking according to a gait of an upper body.